Notes from Main Injector/Recycler Group Meeting Monday, 16 September 2002 Bruce C. Brown Dave Wildman -- Tales of the Famous Dancing Bunches The coupled bunch instabilities which are so prominent in the RF structure of Booster Beam are also seen in the Main Injector. It may be responsible for significant longitudinal emittance growth on $2B acceleration cycles. Dave presented plots from spectrum analysis of the Main Injector beam as seen on the Resistive Wall Monitor (RWM) and also from a selected RF cavity gap monitor. Data from extensive observations on $29 (pbar stacking) cycles and less extensive studies from $2B (150 GeV Protons for Tevatron) cycles were presented. The transparencies from this presentation are to be scanned for your viewing enjoyment at some future date. $29 Cycle Observations Data from the RWM taken with a 20 ms continuous sweep held over the entire acceleration cycle show a number of modes. Some modes are not of interest since they appear in both gaps and the signal coupled in one gap cancels that from the other gap in each cavity. However, important modes are seen at about 127.5 MHz and 224.3 MHz. These are excited by the beam as coupled bunch modes. Looking at the cavity in this fashion also shows these modes. These modes are known and dampers exist. Dave showed a plot from a 1995 paper (with Joe Dey). These are roughly the 3rd and 5th harmonics of the cavity (frequency shifted by the complex cavity geometry). After damping, however, they still have a shunt impedance of about 3 kOhms and a Q ~ 60. Paper available online: http://accelconf.web.cern.ch/AccelConf/p95/ARTICLES/WPP/WPP09.PDF Focusing on each of these modes separately, a zero span measurement of the RWM vs. time in the cycle shows that the 224.3 MHz mode rises except at transition. For the gap monitor, the same measurement shows a signal which builds up slowly. With this measurement, the 127.5 MHz mode shows up on the RWM with a large signal at injection, dropping down early in the cycle before building up again at high energy. Similar signals are seen on the gap monitor. What seeds the instability which drives these modes? An FFT on the first turn signal shows that both of these modes are present. Similarly, an FFT on the last turn in MI before extraction shows again these two modes are prominent. A 1993 Paper (with K. Harkay) on the Booster RF properties shows modes at 83 MHz (for which dampers are operated). This is available via IEEE Xplore (http://ieeexplore.ieee.org/Xplore/DynWel.jsp). Other modes near the 3rd (165 MHz) and 5th (221 MHz) harmonics of the Booster cavities nearly overlap as sidebands. When viewed as sidebands on the 53 MHz signal, the 83 MHz (Booster) and 127.5 MHz (MI) bands overlap. Thus, the incomplete damping of the 83 MHz coupled-bunch mode in the Booster seeds the 127.5 MHz mode in the Main Injector. This was dramatically illustrated by comparing spectra from normal operation with one when the 83 MHz cavity was briefly under repair (not in use). The 127.5 MHz beam signal in the MI was huge. The Booster longitudinal damper provides about a 20 dB (x100 in power) signal reduction. $2B Cycle Observations The measurement on this cycle are less extensive. However comparisons of zero-span measurement at 224 MHz through the $2B cycle were compared with bunch intensities of 310 E9 per bunch and 370 E9 per bunch. Just after transition, the 370 E9 measurement showed a jump in signal of 10 dB. This jump is quite sudden (perhaps faster than the instrument resolution). If confirmed, these measurements will justify a longitudinal damper in the Main Injector. Since the signal grows by x10, we can hope that the power required to control it as it starts will reduce the final signal, thus permitting a smaller damper power. Discussion Bill Foster described briefly the plans to modify the 7.5 MHz Finemet cavity and use a spare RR amplifier to make a broadband damper system to test. He and Wildman propose that a test can be prepared in a couple of weeks if given support including about 4 hours of tunnel time. A complete damper system using Recycler Ring type cavities may be possible in about 6 months. Shekhar observed that tunnel clean-up from the NuMI tunnel contamination (soot which may be corrosive) is essential and will require 1.5 - 2 days of tunnel access.